[jira] Commented: (SLING-431) Unexplained NPE in AbstractSlingRepository.pingAndCheck

2008-05-07 Thread Bertrand Delacretaz (JIRA)

[ 
https://issues.apache.org/jira/browse/SLING-431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12594895#action_12594895
 ] 

Bertrand Delacretaz commented on SLING-431:
---

Note that things worked fine after stopping and restarting the repository 
client bundle.

 Unexplained NPE in AbstractSlingRepository.pingAndCheck
 ---

 Key: SLING-431
 URL: https://issues.apache.org/jira/browse/SLING-431
 Project: Sling
  Issue Type: Bug
Reporter: Bertrand Delacretaz
Priority: Minor

 I've found this trace in the log of my custom launchpad-based Sling webapp:
 Exception in thread Repository Pinger java.lang.NullPointerException
 at 
 org.apache.sling.jcr.base.AbstractSlingRepository.pingAndCheck(AbstractSlingRepository.java:481)
 at 
 org.apache.sling.jcr.base.AbstractSlingRepository.run(AbstractSlingRepository.java:898)
 at java.lang.Thread.run(Thread.java:613)
 This happened at startup, but only once with many successful startups with 
 the same code and configuration.
 The corresponding code line is the log.log call in 
 } catch(RepositoryException re) {
 log.log(LogService.LOG_INFO, pingAndCheck; 
 loginAdministrative failed, re);
 }
 So the only explanation is that log is null, which is impossible looking at 
 the source code: the thread that calls pingAndCheck is created in activate(), 
 which must be called by the OSGi framework after initializing the log member 
 variable:
  /** @scr.reference bind=bindLog unbind=unbindLog */
 private LogService log;
 There is something weird about the log class, in the error log:
 07.05.2008 13:20:39.574 *ERROR* [SCR Component Actor] 
 org.apache.sling.jcr.jackrabbit.client 
 [org.apache.sling.jcr.jackrabbit.client.SlingClientRepository] Cannot load 
 class used as parameter org.osgi.service.log.Lo
 gService (java.lang.ClassNotFoundException: org.osgi.service.log.LogService) 
 java.lang.ClassNotFoundException: org.osgi.service.log.LogService
 at org.apache.felix.framework.Felix.loadBundleClass(Felix.java:1504)
 That happens before the above NPE.
 I won't investigate further for now, but this looks like a weird 
 timing-related bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Commented: (SLING-431) Unexplained NPE in AbstractSlingRepository.pingAndCheck

2008-05-07 Thread Carsten Ziegeler (JIRA)

[ 
https://issues.apache.org/jira/browse/SLING-431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12594904#action_12594904
 ] 

Carsten Ziegeler commented on SLING-431:


I experienced the same strange log entry about the ClassNotFoundException. The 
reason was that my repository client bundle did not have an import for 
org.osgi.service.log. This *should* be generated by the maven plugin, but for 
some strange reasons it did not. Even more strange: the app did run without any 
problems.
So have a look if this import is in the client bundle; if not it should be 
added.

 Unexplained NPE in AbstractSlingRepository.pingAndCheck
 ---

 Key: SLING-431
 URL: https://issues.apache.org/jira/browse/SLING-431
 Project: Sling
  Issue Type: Bug
Reporter: Bertrand Delacretaz
Priority: Minor

 I've found this trace in the log of my custom launchpad-based Sling webapp:
 Exception in thread Repository Pinger java.lang.NullPointerException
 at 
 org.apache.sling.jcr.base.AbstractSlingRepository.pingAndCheck(AbstractSlingRepository.java:481)
 at 
 org.apache.sling.jcr.base.AbstractSlingRepository.run(AbstractSlingRepository.java:898)
 at java.lang.Thread.run(Thread.java:613)
 This happened at startup, but only once with many successful startups with 
 the same code and configuration.
 The corresponding code line is the log.log call in 
 } catch(RepositoryException re) {
 log.log(LogService.LOG_INFO, pingAndCheck; 
 loginAdministrative failed, re);
 }
 So the only explanation is that log is null, which is impossible looking at 
 the source code: the thread that calls pingAndCheck is created in activate(), 
 which must be called by the OSGi framework after initializing the log member 
 variable:
  /** @scr.reference bind=bindLog unbind=unbindLog */
 private LogService log;
 There is something weird about the log class, in the error log:
 07.05.2008 13:20:39.574 *ERROR* [SCR Component Actor] 
 org.apache.sling.jcr.jackrabbit.client 
 [org.apache.sling.jcr.jackrabbit.client.SlingClientRepository] Cannot load 
 class used as parameter org.osgi.service.log.Lo
 gService (java.lang.ClassNotFoundException: org.osgi.service.log.LogService) 
 java.lang.ClassNotFoundException: org.osgi.service.log.LogService
 at org.apache.felix.framework.Felix.loadBundleClass(Felix.java:1504)
 That happens before the above NPE.
 I won't investigate further for now, but this looks like a weird 
 timing-related bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.